Stack Exchange Network

Stack Exchange network consists of 174 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

The design of experiments (DOE) is the study of how to structure an information-gathering exercise where variation is present. In statistics, controlled experiments are usually implied, meaning that study units are randomly assigned to conditions where levels of a variable have been independently ...

0
votes
0answers
5 views

Can prognosis be used to determine patient outcome in Palliative Care intervention?

In this study https://www.ncbi.nlm.nih.gov/pubmed/24559581, palliative care intervention is the independent variable and patient quality of life is the dependent variable. Could the researchers have ...
1
vote
0answers
11 views

How to correctly plan for Missing Data in Longitudinal / Repeated Measures Design?

Using an hierarchical questionnaire for a longitudinal study, I face strict constrains in how many items can be processed by each subject per week. The questionnaire contains 51 Items in total, but ...
0
votes
0answers
6 views

Compare treatment effects across Levels of aggregation

Suppose I am running an experiment to see if a treatment changes the mean weight of a group of people. Note that I am specifically interested in the mean weight: if half the people get heavier, and ...
1
vote
0answers
20 views

Complex neural network design - 250K+ rows / Lot of missing variables

For the past years I have been developing different type of Neural Networks with great success. For the past few weeks I have been working on a (big) project I am kind of struggling with. I hope by ...
0
votes
0answers
11 views

Is it legitimate to design a one-way ANOVA with following post hoc test in mind?

Is it legitimate to design an experiment where I conduct a one-way ANOVA and test for group differences with a post-hoc test? I read on researchgate, that it's bad practice to do so and post-hoc ...
0
votes
0answers
8 views

bayesian optimal design using Rsolnp package

I'v used Rsolnp package to obtain optimal design. I recently realized that this method depends on initial vector of solnp for some models. Why is this happening?
0
votes
0answers
21 views

Specification of mixed model with incomplete factorial design

I have subjects that will be exposed to a treatment. These subjects will be evaluated in four distinct periods (seasons). We distributed these subjects into four groups: Subjects that will NOT be ...
1
vote
2answers
46 views

How to work out p-value if variances of two groups are different?

How do I work out the p value if the variance of my control and experimental group are different? I can't do the t-test anymore. How do I work the p -value out, what method? Also, my two groups are ...
1
vote
0answers
13 views

Randomly select three groups with equal means

I am looking for a method which I can randomly select 100 subjects from each of 3 populations so that each selected group has equal mean body weights. The population data I will be selecting from will ...
1
vote
0answers
25 views

What is the exact probability distribution for a two group experiment?

I recently conducted an experiment testing the effect of a new treatment (compared to a control). I want to compare the effect I observed to the probability distribution for all possible effects. But ...
0
votes
1answer
11 views

interaction between 4-level categorical variable and a continuous moderator

I'm trying to find a way to investigate whether religiosity correlates positively with helping other people, and whether this correlation is different depending on a 4-level manipulation (in my design,...
1
vote
1answer
33 views

2x3 weird design

My experiment tests how frequently people in two different countries use different types of computers. Assume, that there are only three types of computers (desktop, laptop and tablet) . I am ...
0
votes
0answers
8 views

sample size for DoE - root cause for rare event

I have a production process that assemble toys. The process is dependent on 5 different production factors and the process creates defect toys at a rate of 0.05%. I need to design an experiment to ...
0
votes
0answers
14 views

Linear mixed model methodological questions for time series sensor and non sensor data

Methodological question in preparation of an experiment with 4 groups of 10 growth chambers each. It is hypothesized that parasite infestation will affect growth chambers climate, as measured hourly ...
0
votes
0answers
13 views

How to make symmetric model selection with an asymmetric experimental design in a mixed model?

I have $N$ statistical units distributed in $n_t$ groups (each group corresponding to a given time condition), with $n_g$ units per group, s.t $N=n_t n_g$. On each of these statistical units, $n$ ...
0
votes
1answer
24 views

P-value and confidence interval in MVT testing

Context: website A/B testing When calculating P-value and confidence intervals for each variation in an experiment with more than 2 variations (control, variation 1, variation 2 ...) should there be ...
2
votes
0answers
30 views

Why there is no agrement on power / sample size for factorial repeated measures design?

I'm not really sure to what degree this is a statistical question but thought this might be the best place to start. If my questions prove only my ignorance please vote it down and I'll delete the ...
0
votes
0answers
6 views

Name for RCBD with 2nd level of nesting

I'd like to determine the name of a certain statistical design. Specifically, the design involves 5 sites and at each site, 4 blocks are constructed and all treatments are applied within each block. ...
0
votes
0answers
18 views

Experimental design for discrete choice response

I'm designing a survey where people need to choose one of three transportation alternatives. Here are the alternatives, their specific factors/attributes and the levels I intend to use for each: <...
0
votes
0answers
12 views

Study design: small sample sizes with high biased attrition rate during the study duration

I am looking to design an experimental framework for assessing the impact of specific marketing interventions (in this case, consumer participation in an event) on downstream consumer behavior (e.g. ...
0
votes
0answers
10 views

Why does a non-inferiority design require a smaller sample size than a superiority design?

I understand that theoretically, because a non-inferiority design has a wider range of successful outcomes (equivalence or superiority) vs. a superiority design which only has one successful outcome, ...
0
votes
0answers
39 views

How can I estimate the power of a MaxDiff experiment with covariates?

I am running a MaxDiff experiment where there will be four different "arms". Essentially, a respondent will be assigned to one of four conditions, where they will be shown one of three messages (plus ...
0
votes
0answers
20 views

Can I use mean of mean for a/b testing?

I want to design an A/B testing for the independent two-sample t-test. Assume I want to compare the number of items people purchased. group A is the control group, group B is the testing group where ...
0
votes
0answers
18 views

Statistical Response for Rankings when Degree of Difference Matters

Sorry if this question has been addressed recently. I've done some searching around, but have been unable to tell if I am on the right track do to my lack of knowledge on vocabulary in the field. I ...
0
votes
0answers
10 views

Any downsides in giving a participant A and B then compare the results?

Almost all the experimental design studies that I came across randomly separates participants into bucket A or B. Then they study the effects to identify a causal relationship. Are there any ...
0
votes
0answers
19 views

Does an experimental design imply the model to use for analysis?

I am trying to analyse an experiment where the influence of a drug on a variable $Y$ was measured. The subjects have been dispatched into two groups, each group receives a different exposure to the ...
0
votes
0answers
5 views

estimating main effect with two nested factors

Suppose we have two factors sample and temperature and temperature is nested within sample. Is it still possible to estimate the main effect of sample? Thanks very much.
1
vote
0answers
17 views

Constrained design of experiments

I would like to conduct a 2-level fractional factorial experimental design on 8 factors. I used the FrF2 package in R to do so. I have capacity to do roughly 30 experiments, so I selected a resolution ...
1
vote
0answers
29 views

Statistical Design Assistance

I come to you all with a (hopefully) relatively simple question. Sorry for any grammar mistakes, english is not my first language. Background I am creating an experiment that is building off the ...
0
votes
0answers
11 views

Taking into account uncertainty in future values of $S^2$ with continued sampling

This will be a variation on the theme of this question. Suppose you take a sample of small size $n$ from a normally distributed population and you get the Bessel-corrected $S^2= 2.5^2.$ Where I say "...
1
vote
0answers
11 views

How to verify data in a linear fit for a sensor? Experimental vs Expected data

I am trying to characterize a sensor; in which I am trying to obtain the time constant as a function of the pressure applied to the sensor. I fitted a linear trend line into this data. However, I was ...
0
votes
0answers
29 views

How to estimate mixed logit (or random parameter) discrete choice models in R

I have designed a discrete choice experiment to estimate WTP figures for a non-market good within a group where I think preference is heterogeneous. But I am currently struggling to estimate the model....
0
votes
0answers
11 views

Independent Sample T-Test? Or a different statistical analysis? [duplicate]

What kind of statistical analysis should be conducted to compare means of two unrelated groups whereby time (pre and post) is included for each group? For example, wanting to see if one group ...
0
votes
0answers
33 views

Mixed Model ANOVA or Multiple Paired Sample T test?

I'm currently working on my research and this is the design: The aim of the research is to investigate the effectiveness of mindfulness on subjective wellbeing and perceived stress. So, 100 ...
0
votes
0answers
9 views

Trade-offs between several small experiments versus one large experiment

I am applying Design of Experiment principles to create a split-plot experiment that involves several hard-to-change (HTC) factors. I am finding that creating a single experiment covering all the HTC ...
0
votes
0answers
11 views

Determine effect of price change on items

I work for a company that will be changing the prices of a group of items across the chain of stores. My question is, what is the best way to try and determine what the effect is of that price change? ...
1
vote
0answers
16 views

How to identify confounding variables in a design matrix?

Consider for example the following design matrix with columns $[0...k]$: ...
0
votes
0answers
21 views

What is the minimum number of post-baseline time points for a mixed model?

I have 200 case and 200 control patients. There's a baseline time point (n=400) before the intervention began, 4 months later is the first follow=up, one year later is the 2nd follow up, for 1200 data ...
0
votes
0answers
17 views

Experimental design comparing two different drugs

I'm planning a study evaluating the effect (behavior, memory task) of two different drugs at multiple doses (+vehicle control). Ideally I'd like to compare the effects against the control and each ...
0
votes
0answers
14 views

Adding a covariate for the variable dose of a drug

I have a study where subjects (mice) are sleep deprived. Some receive caffeine in their water and others do not and comparisons of their memory are tested over several days. Caffeine consumption, thus,...
0
votes
0answers
28 views

Without running an a/b experiment, how can I show the value of a new product feaure?

I'm an analyst at a consumer facing company working with a product team to understand what value their feature provides our users. There are 2 issues - 1) they ran an a/b test but didn't keep a long ...
1
vote
0answers
11 views

normalizing individual patient reactions to medications in a study

We're trying to launch a study that looks at requirement of pressers, which vasoconstrict, for ICU patients with physical interventions to raise pressure, but some of these patients are under sedation,...
1
vote
0answers
60 views

Experimental Design and Measuring a Rate of Success

Suppose you create an experiment for baseball players where you assign an extra batting coach (treatment) to a group of players and then use a control group of players who did not get the extra ...
1
vote
0answers
25 views

Coefficients in a Randomized Block Design

If, for example, I am running a GLM with Poisson distribution (it could be any distribution) and I have a Randomized Block Design (RBD) [Note: I dont wanna run a GLMM just to put the blocks as a ...
1
vote
0answers
7 views

Unbalanced second-wave recruitment in experimental research

I am working on a consult for a behavioral researcher's ongoing project. The current study uses three experimental conditions (A, B, & C), each with roughly-equal sample sizes (~n) at current. ...
1
vote
1answer
21 views

Experiment where are observations are chi square distributed

In order to introduce the chi-square-distribution I'd like to find a simple experiment where the measurements are chi-squared distributed. Do you know a simple experiment where the measurement ...
4
votes
1answer
34 views

First block not randomised in experimental design. Is it valid?

I've just started a field experiment which my supervisors have already designed and planted for me (It's a maize trial). It is a Randomised Complete Block Design with four replicates, however, the ...
1
vote
0answers
17 views

Confidence Interval for Regression with Input-dependant Variance

I am currently building an optimal designed Experiment. The design will have 20 runs, which will be replicated 10 times, so the total number of runs will be 200. In order to compute the confidence ...
2
votes
0answers
17 views

Experimental Design in A/B Tests

I want to A/B test two separate back-end recommendation logic (A and B) but I'm not sure which method works best (from a statistical standpoint). Here's how my application works: User goes to my ...
2
votes
1answer
61 views

How to infer on a human intelligence task about selecting the best option

I have a human intelligence task test which tries to get the best option from a set of options tested against a control group. The test consist of: Suppose we have a control group of options, lets ...